Abstract:Urban prediction tasks that aim to model the complicated spatial and temporal patterns of urban indicators (such as weather, vehicle charging demand, etc.) for accurate prediction, have been increasingly important in constructing smart cities and accelerating the urbanization process in the modern era. However, most existing works of urban prediction have only concentrated on spatial and temporal correlations, but ignored the effect of distribution shift from spatial and temporal perspectives; this could largely hinder the performance of urban prediction tasks. In order to solve this problem, in this paper, we propose a Shift-Aware Urban Prediction (SAUP) framework to eliminate the inherent shift effect among spatial-temporal urban time series data. Specifically, SAUP starts with a Shift Elimination Module, built upon our proposed Spatial-Temporal Attention Flows (STAF) composed of invertible attentions and coupling layers of normalizing flows in order to transform the raw shifted data into a unified distribution to remove the spatiotemporal shift. After the shift effect is eliminated, the Correlation Processing Module of SAUP further captures the core correlations to learn spatiotemporal dependencies, in which topological correlations and geographic correlations are jointly learned by GCN and CNN based on pre-defined graphs and extracted POI information. In addition, SAUP includes a model-agnostic Forecasting Module, which can be employed as any forecasting architecture to accomplish the predictions. To recover the raw distribution information, the output of the Forecasting Module is further taken for the inverse transformation of the Shift Elimination Module to produce the final forecasts. We have conducted extensive experiments in the SAUP framework, coupled with six state-of-the-art spatiotemporal forecasting models on two real-world datasets. Experimental results have demonstrated the consistent improvements of SAUP over the baseline algorithms.

Enhanced Urban Region Profiling with Adversarial Self-Supervised Learning for Robust Forecasting and Security

Urban surface classification using self-supervised domain adaptive deep learning models and its application in urban environment studies

ESCORT: Fine-Grained Urban Crime Risk Inference Leveraging Heterogeneous Open Data

Unsupervised Domain Adaptation for Crime Risk Prediction Across Cities

A human-machine adversarial scoring framework for urban perception assessment using street-view images

MSTEM: Masked Spatiotemporal Event Series Modeling for Urban Undisciplined Events Forecasting

Dynamic road crime risk prediction with urban open data

Boosting Urban Prediction via Addressing Spatial-Temporal Distribution Shift

Adversarial Attack Against Urban Scene Segmentation for Autonomous Vehicles

Spatio-Temporal Analysis of Urban Crime Leveraging Multisource Crowdsensed Data

Robust Spatiotemporal Traffic Forecasting with Reinforced Dynamic Adversarial Training

Self-Supervised Likelihood Estimation with Energy Guidance for Anomaly Segmentation in Urban Scenes

UP-Diff: Latent Diffusion Model for Remote Sensing Urban Prediction

Meteorology -Assisted Spatio-Temporal Graph Network for Uncivilized Urban Fvent Prediction

Meteorology-Assisted Spatio-Temporal Graph Network for Uncivilized Urban Event Prediction.

Spatial-Temporal Meta-path Guided Explainable Crime Prediction

Adversarial autoencoders and adversarial LSTM for improved forecasts of urban air pollution simulations

Urban Region Profiling via A Multi-Graph Representation Learning Framework

Enhance Sample Efficiency and Robustness of End-to-end Urban Autonomous Driving via Semantic Masked World Model

CityNeuro: Towards Location and Time Prediction for Urban Abnormal Events

Urban Perception: Sensing Cities Via a Deep Interactive Multi-task Learning Framework